The hypothesis to be tested in this project is that the identification of chemopreventive agents in skin cancer will be facilitated by the discovery of new molecular targets in pre-invasive and early invasive lesions. Underlying this hypothesis is the recognition that genome-scale phenotyping (combining gene expression with gene copy number) of skin cancers will allow identification of sub-types sharing distinct molecular etiologies across the set of tumors under study. Our efforts involve the use of a series of computational approaches developed in our laboratory to study sets of genes that sharply differentiate recognizable subtypes and will implicate genes and/or proteins likely to be crucial to the maintenance and progression of these disorders. We will detail an experimental approach that will address important clinically relevant deficiencies (both diagnostic and prevention-related problems) in the area of cutaneous oncology, and provide new opportunities for positively impacting these increasingly important disorders. Four specific aims are designed to identify, validate and provide initial pre-clinical testing of targets for chemopreventive agents. Molecular targets identified in human samples will be candidates to evaluate against molecular targets from two transgenic mouse models of melanoma (UV-induced and spontaneously developing), as well as against UVA and UVB-induced murine models (linking this project to Projects I and II). Taking into account the known biology of a targeted gene, its correlation with survival and or progression and with cellular phenotypes such as proliferation, differentiation, cell death or invasion will provide a rational approach to identify targets for clinical chemopreventives (linking this project to Project IV). A secondary benefit of this project will be the generation of data useful in defining a molecular-based taxonomy of actinic keratoses (AK) and more progressed lesions for SCC, and dysplastic nevi (DN) and more progressed lesions for melanoma. This is a unique project in cutaneous oncology that combines multiple genomic strategies and relevant animal models, within a world-class clinical translational research program